EROTIC DREAMS AND NIGHTMARES FROMANTIQUITY TO THE PRESENT*

Charles Stewart

University College London

The history of erotic dreams, nightmares, and erotic nightmares offers a valuable oppor-tunity to study how such dreams tested Western ideas about the self, desire, and self-control. Like Foucault, I find it more productive to analyse these dreams, and the strugglesto introject them, as sites of self-making rather than of repression. Erotic dreams andnightmares have been inflected by various historical strategies of self-making, themselvesproduced by different regimes of knowledge such as Christian asceticism, medicine, orphilosophy. Erotic nightmares still proliferate today in reports of alien abductions.A reasonfor this historical tenacity has been the ease with which the affective sensations of theerotic nightmare – terror and sexual arousal – have jumped between genres as various asmonastic handbooks, medieval folk-tales, gothic fiction, and personal dreams. This studydemonstrates the importance of historical perspective for the ability to identify andunderstand culturally elaborated (‘culture-bound’) syndromes.

Men make their own history, but they do not make it just as they please; they donot make it under circumstances chosen by themselves, but under circumstancesdirectly encountered, given and transmitted from the past.The tradition of all thedead generations weighs like a nightmare on the brain of the living (Die Traditionaller toten Geschlechter lastet wie ein Alp auf dem Gehirne der Lebenden).

Marx, The eighteenth brumaire of Louis Bonaparte

Erotic dreams have raised perennial questions about the boundaries of the selfand the individual’s ability to control and produce this self. Do erotic dreamsresult from divine intercession, an immoral life, or recent memories? Are theyproducts of the self for which the individual dreamer may be held respon-sible? Or are they determined by a force majeure such as original sin, or humanphysiology?

The answers that various societies supply to these questions no doubt con-dition the ways in which people in different cultures and historical periodsreact to their experiences of erotic dreams. The Hadza of northern Tanzaniapublicly marked a boy’s first nocturnal emission by decorating him with beadsin exactly the same way as they decorated a girl with beads at the time ofher first menstruation. Both occasions were unequivocally positive and cele-bratory (Woodburn, pers. comm.; see also Woodburn 1964: 269, 303, table 19,ill. 14). By contrast, the monks addressed in the fifth-century CE writings ofJohn Cassian were instructed that:

© Royal Anthropological Institute 2002.J. Roy. anthrop. Inst. (N.S.) 8, 279-309

* Malinowski Memorial Lecture, 1998

It [an emission] is a sign of some sickness hidden inside, something hidden in the inmostfibres of the soul, something that night-time has not produced anew but rather hasbrought to the surface of the skin by means of sleep’s restorative powers. It [night-time]exposes the hidden fibres of the agitations that we have collected by feasting on harmfulthoughts all day long (Cassian, Institutions, 6.11, trans. in Brakke 1995).

Granted the value or danger accorded to erotic dreams in different soci-eties, it is not surprising that vastly different practical techniques have beenformulated cross-culturally to cultivate intended results. In order to fend offerotic dreams Graeco-Roman doctors variously recommended sleeping onone’s side, excluding warming foods from one’s diet, sleeping with a lead platein contact with one’s testicles, or having intercourse in the dark so as to avoidmentally registering lust-provoking visual images that could later recur duringsleep (Foucault 1986: 137 ff.). By contrast, among the Umeda of Papua NewGuinea a hunter intentionally slept on a net-bag scented with magic pig-hunting perfume (oktesap) in hopes of receiving the erotic dream that pre-saged a successful hunting expedition. Such erotic dreams held out the promiseof real sexual consummation, which often followed after a kill was made (Gell1977: 33).

Just how, and how successfully, such culturally prescribed bodily practicesactually affect the subjective phenomenological apperception of dreams is adebated issue to which I shall return near the end of this article. I beginsimply with the observation that the social construction of erotic dreams man-ifestly varies, cross-culturally and diachronically. In examining the historicalnightmarization of the erotic dream within European history, I present a corol-lary to Foucault’s thesis (1978; 1985; 1986) that sexuality increasingly becamea site of ‘subjectivation’ (assujettissement) in the West. By this neologism hemeant that sexual desire became the indicator of the truth about one’s self,and thus a fundamental constituent of one’s subjectivity. At the same time thissexuality was also a conduit for subjugation by social forces such as the Churchor medical science that instructed how desire should be regulated. In volumestwo and three of The history of sexuality Foucault frequently considered theevidence of dreams, particularly erotic dreams, and the challenges they posedto images of self-control in antiquity. Here, I track dreams into the Christianperiod and through the Middle Ages, thus proceeding into areas that Foucaultwould have covered more extensively had he lived to fulfil his original planfor the The history of sexuality.1

Medieval conceptions of erotic dreams, nocturnal emissions, and the dangersthey posed to monks, clergy, and the Christian laity have been the subject ofsome excellent recent studies (Brakke 1995; Elliott 1999). And scholars ofancient Greece have paid increasing attention to erotic dreams and the codesfor interpreting them (Grottanelli 1999;Winkler 1990a). But there have beenalmost no studies of the nightmare in antiquity save for two works from theearly twentieth century.The German classical scholar Wilhelm Roscher tracedthe nightmare from ancient Greece through the Renaissance in a work entitled Ephialtes (1900).The other study, Ernest Jones’s On the nightmare (1971[1931]), was originally published as two articles around 1910. Both Roscherand Jones took for granted an erotic dream-nightmare complex already inclassical antiquity.

280 CHARLES STEWART

In my view, the erotic dream and the nightmare were not systematicallyconjoined until the early Christian period, when the control of inner cupid-ity became a salient diagnostic of spiritual progress. At this time potentiallypleasurable erotic dreams were demonized and fused with a particular, pre-existing conception of the asphyxiating nightmare to give rise to a nightmare-erotic dream complex.

Since the advent of psychoanalysis, the concept of ‘repression’ has oftenbeen applied to account for the erotic nightmare. Take for example the fol-lowing orthodox Freudian statement by Jones:

The psycho-analytical conclusion about nightmares is that the causative sexual wish – soevident in most erotic dreams – is subject to an exceptional degree of repression, thereason for this being that they always originate in an incestuous wish. The attackinganimal, demon, or vague pressure really represents the parent. It is probable that theextreme classical form of nightmare occurs only in persons with a considerable masochis-tic element in their constitution ( Jones 1974 [1932]: 113).

Terror and eros, for Jones, stood in tense complementary relationship andformed a continuum from exclusively erotic dreams at one pole, through gra-dations of erotic nightmares, to the other extreme where there was only terror(1971 [1931]: 42; 1974 [1932]: 111). All erotic dreams were incipient night-mares. Furthermore, since the Oedipus complex was universal, the same psy-chological tensions would generate erotic nightmares cross-culturally andtrans-historically.

Of course Malinowski fought the battle over the universality of the Oedipuscomplex against the very same Ernest Jones who wrote On the nightmare. Manypeople assume that Malinowski had little sympathy for psychoanalysis, but headmired much in Freud’s work (Malinowski 1962 [1923]; Stocking 1986). InSex and repression in savage society (published in 1927) he certainly embracedthe psychoanalytic concept of repression. But this seems to have been the psychoanalytic idea that he least understood. In Sex and repression he beganby asserting that because the Trobrianders were unrepressed they did not havedreams at all (1965: 89). He then proceeded to argue (contradictorily) that ifa dream of sex with one’s sister was more disturbing than a dream of sex withone’s mother, then this proved the existence of a matrilineal rather than anOedipus complex (1965: 91).

Repression remains an area of misunderstanding between psychology andanthropology to this day. Few studies have attempted to clarify the viabilityof this term for anthropological research.2 Below, I examine the psychoana-lytic concept of repression by contrasting it with conscious and intentionalacts of exclusion. Much of my historical evidence reveals people voluntarilyattempting to repel unwanted thoughts. In Freud’s terms they are ‘suppress-ing’ rather than ‘repressing’.3 Intentional self-making thus generated the eroticnightmare in many cases, although it must be granted that overarching reli-gious, political, scientific, or other frameworks outlined the sorts of selves thatpeople should ideally craft and the techniques to be used. The erotic night-mare consequently developed in the West as a culturally interpreted set ofsymptoms, verging on what anthropologists used to call a ‘culture-bound syn-drome’ (Low 1985). The diachronic examination of the erotic nightmare

CHARLES STEWART 281

complex offered here might usefully remind those studying other culturallyelaborated syndromes such as latah, nervios, or koro that these complexes arenot permanent cultural badges, but symptoms of/in transition.4 Each has itsown current trajectory and velocity, which require historical contextualiza-tion. Thus, whereas for Marx history was like a nightmare, in my view thenightmare has a history worth exploring for what it might reveal about suc-cessive Western conceptualizations of terror, sexual desire, and the self.

Erotic dreams and nightmares

Clearly we need to decide on the definition of both erotic dreams and night-mares before we can proceed to identify their combination.The word ‘night-mare’ can be used to describe any unpleasant or terrifying dream that disturbssleep, and by the end of this article I shall be using the term in this broadsense. I begin, however, in the ancient historical sections, with a morerestricted conception of nightmare, namely, dream visions where the sleeperfeels a weight on his or her chest and has accompanying sensations of dys-pnoea, paralysis, and dread. Sufferers frequently stated that a person, an animal,or a demonic being caused these feelings by sitting (or jumping up and down)on their chest. The Swiss artist John Henry Fuseli’s painting The nightmare,first exhibited in 1782 at the Royal Academy in London, gave this body oftraditional ideas a compelling representation. A hairy demon perches atop avoluptuous woman laid out asleep on her bed as a wall-eyed horse pokes itshead through the velvet drapes in the background.5 (Fig. 1.)

There are grounds for calling this the ‘classic nightmare’ because the imageof demonic pressure informs the current vocabulary for the phenomenon ina number of languages. In my epigraph, Marx’s word for the nightmare is Alp,meaning ‘elf ’, and it draws on the two usual German words for nightmare:Alptraum (elf dream) and Alpdruck (elf pressure). The words for ‘nightmare’ inother languages also convey this image: French cauchemar etymologically means‘oppressive fiend’; the mar(e) component meant variously a water monster,vampire, or just a ghost at various stages and in different registers of French(Gamillsches 1969). According to Dr Johnson’s Dictionary the mara was ‘aspirit that, in the Northern mythology, was said to torment or suffocate sleep-ers’ (Frayling 1996: 8). Our word ‘nightmare’ comes from this Scandinaviansource (Tillhagen 1960).

Numerous distinctions also need to be made in respect to the concept of‘erotic dream’. The criterion of erotic requires that the dreamer feel sexuallystimulated, yet this is often difficult to establish in relation to historical dreamsand their dreamers. In the relationship between dream imagery and eroticresponse, virtually any configuration is possible. In 95 per cent of cases penileerection in males and analogous genital excitation in females accompany REMsleep (Greenhouse 1974).Almost every dream is thus built upon erotic impulses.

Greek antiquity

In ancient Greek dream books sexuality displayed in dreams was usually read as diagnostic of more important issues (for the ancients) such as wealth

282 CHARLES STEWART

CH

AR

LES ST

EW

AR

T283

Figure 1. J.H. Fuseli, The nightmare (1781). Founders Society Purchase with funds from Mr. and Mrs. Bert L. Smoklerand Mr. and Mrs. Lawrence A. Fleischman. ©2000 The Detroit Institute of Art.

and social status (Foucault 1986: 33; Winkler 1990a: 27). For the second-century CE dream interpreter Artemidorus, the dream of sex with one’smother, for example, was not problematic, but rather a good dream for poli-ticians. This was because the mother represented one’s native country, and to make love is to govern the obedient and willing body of one’s partner.The dreamer would thus control the affairs of the city (Artemidorus, Inter-pretation of dreams, 1.79). Hippias, a Greek traitor serving as the Persians’ guidein the landing at Marathon, dreamt of sleeping with his mother and inter-preted this to mean that he would return to Athens and recover power(Herodotus, History, 6.107).

Artemidorus divided dreams into two main categories: enypnia and oneiroi.Enypnia directly expressed current desires or bodily states. A hungry mandreams of eating, a lover of his/her beloved (Artemidorus, Interpretation ofdreams, 1.1). As they predicted nothing, such dreams did not require the ser-vices of a dream interpreter. Oneiroi, on the other hand, were prophetic dreamsproduced by the soul, or possibly sent by the gods. There could thus be twokinds of sexual dream. One type merely indicated that the dreamer was ‘inthe mood’, while the other employed sexual scenarios to convey prognosticinformation about entirely different matters. Artemidorus dispensed withenypnia in a couple of paragraphs, while devoting four chapters to sexualoneiroi.

As sexual oneiroi were not really about sex it is quite possible that they did not involve an erotic response in the dreamer. This arouses the suspicionthat numerous manifestly sexual dreams from other contexts were similarlyunerotic. Consider, for example, the following inscription from the healingshrine of Asclepius at Epidaurus recorded in the fourth century BCE tomemorialize a dream by a female pilgrim from Messene: ‘Sleeping here shesaw a dream. It seemed to her the god came bringing a snake creeping besidehim and she had sex with it. And from this children were born to her withina year, twin boys’ (LiDonnici 1995: 115).The characteristics of the nightmareare arguably there alongside the erotic content. But perhaps the sex is nothingmore than the metaphorical expression of contact with the god. In the absenceof a dreamer’s comment it is difficult to know which dreams were erotic orfrightening, or both at the same time.We must look beyond myths and dream-books to ancient Greek conceptions of self and emotion.

In Aristotle’s system of thought dreaming was situated in relation to per-ception.The difference was that in normal perception one submitted ‘appear-ances’ to judgement in order to decide whether something was or was notthe case.The sun may appear to be the size of a penny but judgement enablesone to decide that it is not really so small. During sleep, however, the capa-city to judge is suspended with the result that the images presented to thesleeper’s mind are automatically accepted as valid.6 The only exceptions to thisrule were rare instances when one somehow attained consciousness duringthe dream and judged a given appearance false (Aristotle, On dreams, 462a5).In the majority of cases, however, people reacted to dream images as if theyhad judged them and accepted them as real. This accounts for people’s fullemotional response within dreams because emotions inevitably ensue uponjudgements. If we decide that a situation is frightening, or erotic, then we arefrightened, or aroused.

284 CHARLES STEWART

Aristotle’s teacher, Plato, understood dreams as more closely involved withindividual character. He focused on dreams as a facet of his overall study ofpolitics, since he considered that the credentials to govern others were firstdeveloped and proved in relation to oneself.7 In Plato’s view the soul ( psykhe)was comprised of three parts: the rational mind (nous), high spirits (thymos, forexample, anger, joy, courage) and the appetitive desires for food, drink and sex(epithymia) (Republic 441a). Self-mastery involved regulating these three com-ponents and integrating them under the command of reason. One recom-mended technique involved forging a strategic alliance between reason andthe high spirits so that the appetitive desires were double-teamed. The edu-cation of the young through dancing and verse-recitation exemplified thisstrategy since the chorus songs fostered the growth of reason and knowledgein the intellect, while the music and dance calmed the high spirits withharmony and rhythm (Plato, Republic 441e).

All humans possessed the ability to control their appetites but not all ofthem exercised it. In any case, sleep presented a dangerous moment. Whilereason slumbered the way was open for the savage part of the soul to breakfree and express itself, especially if the person had indulged in excessive eatingand drinking.

I’m sure you’re aware of how in these circumstances nothing is too outrageous: a personacts as if he were totally lacking in moral principle and unhampered by intelligence. Inhis dreams, he doesn’t stop at trying to have sex with his mother and with anyone oranything else – man, beast, or god (Plato, Republic 571c).

Whereas men could overcome or at least moderate their responses to desiresby internal mental effort, women were conceived to be fundamentally help-less victims of their insatiable sexual appetites (Dean-Jones 1992). Accordingto ancient medical thought, female physiology precluded reason being exer-cised by women in the same way as by men.This was because the womb, theseat of her sexual appetite, was not under a woman’s conscious control. Itcould move around in her body, even up to the head, where it could ‘stiflethose organs in which consciousness was thought to lie’ (Dean-Jones 1992:78). Thus women had no chance to succeed or fail in the fundamentallyethical arena of relating to their own sexual desire.They were denied the pos-sibility of acting as moral agents (Dean-Jones 1992: 86).

The erotic nightmare in antiquity?

The ancient Greeks did not apparently expend much effort in analysing fright-ening dreams. Our best evidence for the ancient conception of the nightmareis the term itself, ephialtes. Etymologically it seems to mean ‘to jump on topof ’ (Chantraine 1977).The first-century CE physician Themison of Laodicaeareportedly coined another name for the nightmare, pnigalion ‘strangler, throt-tler’.8 This is about all we have to go on.

Did it also have an erotic component? Roscher thought it did, on accountof the identification between the god Pan and Ephialtes, but there was noclear indication that these Pan nightmares possessed an erotic dimension untilArtemidorus, who wrote that:

CHARLES STEWART 285

Ephialtes is identified with Pan but he has a different meaning. If he oppresses or weighsa man down without speaking, it signifies tribulations and distress. But whatever he saysupon interrogation is true. If he gives someone something or has sexual intercourse withsomeone, it foretells great profit, especially if he does not weigh that person down(Artemidorus, Interpretation of dreams, 2.37).

Here, at last, all of the components of the erotic nightmare are present, butwe are well into the Roman imperial period and Christianity had alreadybegun to spread.

This, then, is the evidence for the erotic nightmare complex in antiquity.There is no reason to suppose that such nightmares would necessarily havebeen systematically recorded, but this is really very little. I do not wish toargue that the ancient Greeks never experienced erotic nightmares.Clytemnestra’s dream of a snake biting her breast (in Aeschylus’ Libation bearers,523-54), or the general imagery of Euripides’ Bacchae, suggest the contrary.But the ancient Greeks did not apparently develop a discourse about the eroticnightmare.They did not create a nosological category for it, or any other sortof denominated cultural classification. Erotic dreams and nightmares wereapparently treated as two distinct phenomena.

This situation changed with the coinage of the Latin word incubus (a demonthat ‘lies/sleeps upon’ the dreamer) around the beginning of the CommonEra.The first meaning of incubus was simply ‘nightmare’ and it may have beena straightforward attempt to translate the Greek ephialtes. In rendering the ideaof ‘jumping upon’ into Latin, however, the translators enmeshed it in a matrixof words that contained clear sexual connotations – for example, concumbere‘to sleep (with)’ and concubinus ‘concubine’. The erotic overtones in the wordincubus would ramify in Christian conceptions of erotic dreams and night-mares, furnishing a good example of ‘semantic contagion’ (Hacking 1995a:238). The coinage, several centuries later, of the term succubus to denote ademon that lay beneath the sleeper indicates how productive the erotic night-mare concept became.This proliferation of linguistic categorizations contrastswith the paucity of terms in Greek of the ancient period.

A marble relief (Fig. 2), datable only very generally to between the secondcentury BCE and second century CE, encapsulates the uncertainties over thetransition to the erotic nightmare during this same broad stretch of time.9 Onthe one hand, the relief reveals a public acceptance of erotic imagery charac-teristic of Greek antiquity.A sleeping shepherd, his staff and bagpipe cast downby his side, is straddled by a winged figure – the symbol of the dream or itscontent in ancient Greek iconography (Boardman & La Rocca 1978: 159).On the other hand, the female figure, with her wings, and webbed feet, mightbe interpreted as a type of demon, possibly a siren. Pleasant dream, or terri-fying assault of the noonday demon? We cannot know for certain what thisshepherd is experiencing.

Refinements in moderation

Before coming to Christianity I would like to consider the later life of theclassical idea of moderation (sophrosyne) as it was developed in the medicaltradition and in schools of philosophy such as Stoicism. For those practising

286 CHARLES STEWART

moderation sex was not problematic, so long as one kept the whole body inbalance. Indulgence of the appetites was negotiable according to the age andgender of a person and the season of the year. Imbalances could be correctedby medically prescribed diet and exercise regimens. Sexual activity was onlya concern if it took uncontrollable forms.

Some ancient doctors understood ‘gonorrhoea’ to be an involuntary emis-sion of semen, and their term for this ailment, meaning literally ‘the flow ofseed’, remains with us to this day. Nocturnal emissions were considered avariant of gonorrhoea, and in his survey of acute and chronic diseases CaeliusAurelianus contrasted the two.10 Gonorrhoea could occur any time, without

CHARLES STEWART 287

Figure 2. Old man and a siren. Fragment of a marble relief. Courtesy, Museum of Fine Arts,Boston (08.34c). Reproduced with permission. ©Museum of Fine Arts. Boston. Allrights reserved.

imagery, while nocturnal emissions occurred only during sleep and as a con-sequence of imagining sexual intercourse through ‘unreal images’ (inanibus visisconcubitum fingat) (On chronic diseases, 5.71.82). Unlike gonorrhoea, nocturnalemissions did not necessarily constitute an illness. They simply resulted fromdesire, which could arise either through regular sexual practice or throughprolonged continence.

Caelius Aurelianus devoted a separate chapter to the nightmare (incubo) andit is found in a completely different part of the treatise on chronic diseases,near to the sections on epilepsy and madness. Like the nocturnal emission, heconsidered it an illness only if it occurred chronically. He mentions the factthat ‘[s]ome are driven into such a state by unreal images that they imaginethey see the attacker urging them to satisfy a shameful lust’ (On acute diseases,1.3.56 ff., my trans.). He thus described both the (erotic) nightmare and theerotic dream as physical responses to imaginings, but did not consider themnecessarily to be signs of ill health.

The possibility that people might be able to respond differently to unrealfigments of the imagination resonates with Stoic ideas developed during thelast centuries BCE. Chrysippus emphasized the difference between impres-sions (phantasiai) resulting from the perception of real physical objects, and fig-ments (phantasmata) produced by the imagination and occurring especially ‘inpeople who are melancholic and mad’ (Chrysippus, in Aëtius, Placita, 4.12.1-5: Long & Sedley 1987: 237). According to the Stoics, appetite, fear, distress,and pleasure comprised primary emotions – states not produced, but only suffered by the mind (Stobaeus 2.88, 8-90, 6: Long & Sedley 1987: 411).The term for emotion, ‘pathos’ in Greek, could mean ‘passion, affection’ as wellas a passive ‘suffering’. Active control or passive submission to the emotionswas precisely the issue. The early Stoics held that all passions were the resultsof judgements, and thus could be modified,11 and their goal was to reach astate of apatheia (impassivity) in which one had eradicated uncontrolled emo-tional responses entirely, and thus eliminated passive suffering from one’s life.If this were successfully achieved one could be happy, while those whoneglected actively to confront the passions were – in latter-day terms –pathetic.

The mixed dream

Early Christian preachers such as Justin Martyr assimilated all of the pagangods to ‘demons’ under the control of the Devil (Pagels 1988: 42). Accordingto pagan cosmology, demons were not intrinsically evil, but they were bidd-able. The magical papyri of the last centuries BCE and first centuries CEreveal how people sought, through ritual incantations, to command demonsto carry dreams to others. In one particular example, a man named Hermeiasexhorts the demons to cause his unresponsive object of desire to lust for him,even when she is ‘drinking, working, conversing, sleeping, dreaming, havingan orgasm in her dreams, until she is scourged by you and comes desiringme’.12

Granted the prevailing association of demons with dreams in popularthought, Christians were counselled to distrust their sleeping visions as pos-

288 CHARLES STEWART

sibly satanic. Dreams thus came to be placed squarely on the negative side ofa morally polarized universe. John Climacus, whose Ladder of divine ascent syn-thesized the ascetic tradition and became a handbook for monks, wrote:‘Devilsoften take on the appearance of angels of light or martyrs and they appear tous in sleep and talk to us … And if we start to believe in the devils of ourdreams, then we will be their playthings when we are also awake’ (Climacus,Ladder, 3).

Beginning with Tertullian, the Church Fathers held that dreams could comevariously from God, the Devil or the Soul (Tertullian, On the soul, 47). Thistripartite scheme was apparently adapted from pre-Christian philosophical traditions.A look at the third-century BCE Alexandrian physician Herophilus’classification of dreams reveals that the erotic dream figured centrally in thetransition from paganism to Christianity:

Herophilus says that some dreams are inspired by a god and arise by necessity, whileothers are natural ones and arise when the soul forms for itself an image (eidolon) of whatis to its own advantage and of what will happen next; and still others are mixed (synkra-matikoi ) and arise spontaneously (ek tou automatou) according to the impact of the images,whenever we see what we wish, as happens in the case of those who in their sleep makelove to the women they love.13

The interesting part of this scheme is the third, or mixed category. In so faras people see what they inwardly desire in these dreams, they seem identicalto enypnia – the physical state dreams discussed earlier.14 Yet, this identifica-tion cannot be correct, since Herophilus pointedly differentiates them fromthe category of dreams produced exclusively by the soul. Mixed dreams havean exogenous element; they result from outside forces – the impact of imageson the sleeper. These images happen to coincide with internal desires.

Herophilus’ mixed dream, with its ready erotic exemplification, corre-sponded to the demonic dream in the Christian tripartite system (von Staden1989: 310). Early ascetic theories of human nature and psychology reveal howmonks understood demons to inspire erotic dreams.These accounts, presentedby writers such as Evagrius and Cassian, possibly illuminate what Herophilusintended by the mixed dream. Certainly they take us deeper into the genesisof the erotic nightmare.

For Evagrius, who became a monk in Egypt around 382 CE, demons could manipulate an individual’s previously acquired, emotionally charged momories to excite the passions, and set sinful thoughts in train. Thus evilthoughts were simultaneously exogenous and endogenous; demons activatedwhat was already there. Evagrius conceded that disturbing thoughts wouldinevitably occur, even in the course of monastic life – such thoughts werepart of the human condition. Sin set in only if one mentally entertained athought for too long. As he expressed it: ‘It is not up to us whether evilthoughts might trouble the soul or leave it in peace. What does depend onus is whether they linger or not, and whether they set the passions in motionor not’ (Praktikos, 6). The goal was inner stillness, which Evagrius referred toby the familiar Stoic term, apatheia (Guillaumont 1971: 98 ff.).

Evagrius named eight primary demons, the model for what would becomethe ‘seven deadly sins’ in Western Christianity. Each of these demons normally

CHARLES STEWART 289

attacked only one of the two vulnerable parts of the soul, the high-spiritedor the sensual. Predictably, the demon of fornication ( porneia) attacked thesensual part of the soul. According to Evagrius it

compels one to desire ‘remarkable’ bodies; it violently attacks those living in abstinencein order to cause them to quit, convinced they will amount to nothing. And, soiling thesoul, it inclines it to ‘those acts’ [obscene acts]. It causes monks to speak and hear things,as if some object were visible and present (Praktikos, 8).

As this passage shows, the battle with demons spilled over into the realmof dreams and (other) hallucinations where the power of the will to resistdemons was weakest. Although demons could provoke erotic dreams andnightmares, these were normally two distinct types of dream (Praktikos, 21, 22,54). The phenomenon of an erotic nightmare required a fusion of demonicdomains that contravened the normal division. In such dreams the sensual partof the soul joined forces with the irascible to overwhelm the intellect. It wasthe opposite of Plato’s ideal scenario of self-mastery where the intellect andthe high spirits co-operated to overpower the appetitive part. In Evagrius’ psy-chology the erotic nightmare was not excluded but rather given a powerfultheorization. It was the exception that confirmed the rule.15 The erotic dreamwas a mixed dream then, not only because external demons aroused internalthoughts, but also because it simultaneously affected the two parts of the soul.

If dreams were, indeed, controllable, then anyone who experienced an eroticdream was potentially culpable. John Cassian excused nocturnal emissions ifthey occurred to someone with a full stomach (Cassian, Conferences, 12.2). Insuch cases they were a simple physical fact of the body, and he allowed thatit was ‘natural’ for emissions to occur as often as every two months, althoughthree times a year was a more acceptable frequency (Cassian, Institutions, 6.20;Conferences, 2.23).

The sinfulness of erotic dreams and nocturnal emissions continued to be atopic of debate in ascetic ‘anthropology’ – as patristic theories of human natureand psychology are sometimes known – throughout the Middle Ages in boththe Eastern and Western Churches (Elliott 1999; Fögen 1998). Excusable nocturnal emissions became sinful erotic dreams if one entertained them,allowed them to linger, and, most importantly, if one consented to them(Elliott 1999: 20). The way to fight the images and sensations of the mixeddream was to sever them with the knife of the will, withholding assent sothat externally instigated images did not connect with bodily passions. Noc-turnal emissions unaccompanied by visual imagery indicated spiritual progress(Evagrius, Praktikos, 55; Angelidi forthcoming).

From the monastery to the world

The account developed to this point presents the views of learned texts representing the ideas and practices of elite, free men in antiquity and a narrowsubsection of monks and high clerics in the early Christian period. Their practices of self-cultivation may not have been shared by very many of theircontemporaries, but their influence on subsequent generations has been enormous. If the ancient Greek ethic of self-moderation was explicitly elitist,

290 CHARLES STEWART

the Christian ethic became increasingly unified in conception and intendedfor all – men and women, young and old alike. I turn now to consider howthe ascetically influenced Christian ethic of self diffused to the population atlarge and how the Christian laity was conditioned to view the erotic dreamas dangerous and nightmarish.

In popular vocabulary the word incubus, as we saw, gave people a ready labelfor the erotic nightmare.16 In the wake of Augustine’s writings about con-cupiscence and original sin, the general term for demonic interference in adream, ‘inlusio’ (illusion), came to have automatically erotic overtones in suc-ceeding generations (Elliott 1999: 20). Likewise, the term ‘phantasma’, whichAristotle had used interchangeably with phantasia to mean a ‘mental percep-tion, image, or representation’, came to mean a distorted – usually by demons– mental representation (Schmitt 1999: 278). If normal sensory perceptionswere like water that flowed through a person, then memories could be likenedto water that was stored and which remained clear. Phantasmata, on the otherhand, were like stagnant water that had become cloudy, rank, and overgrownwith algae.

How did these developments affect popular views of these matters? Cer-tainly the laity were not expected to live up to the ascetic standards of themonks – this would have meant the extinction of Christian society – norwere they necessarily concerned by, or even able to comprehend, the high-flown arguments of theology. People in the world no doubt continued to haveextra-marital sexual relations, dreams, erotic dreams, nocturnal emissions, andnightmares. But the Church did make attempts to regulate these phenomena.Early penitential books such as the Irish penitential of Cummean, composedin the seventh century after the model of Cassian’s rules for monks,represented one such effort. This penitential is notable for its comprehensivedistinctions among erotic deeds and thoughts.

He who merely desires in his mind to commit fornication, but is not able, shall dopenance for one year … He who is willingly polluted during sleep, shall rise and singnine psalms in order, kneeling … He who desires to sin during sleep, or is unintention-ally polluted, fifteen psalms; he who sins but is not polluted, twenty-four (Bieler 1963:115; Asad 1993: 101).

The dissemination of prayer formulas comprised another area for ascetic influ-ence on the development of mainstream Christianity. The expanding practiceof bedtime prayers is of particular interest here (Le Goff 1988: 225). Early inthe fifth century CE, Prudentius composed a hymn before sleep that includedthe following lines: ‘If a man’s stains of guilty conduct are few and far between,him the clear and flashing light teaches secret things; but he who has pollutedand befouled his heart with sins is the sport of many a fear and sees fright-ful visions’ (Daily round 6.49). And it concluded with the following exorcis-tic entreaty: ‘The cross drives out every sin; before the cross darkness fleesaway; consecrated with this sign, the spirit cannot be unquiet.Away, away withthe monstrosities of rambling dreams! Away with the deceiver and his persis-tent guile!’ (Daily round 6.133).

Between the fifth and thirteenth centuries, the Church’s mode of elicitingand forgiving lay sins altered. Initially, there was the brutally demanding office

CHARLES STEWART 291

of penance in which the penitent was excluded from the worshipping com-munity (Asad 1993: 100). This person’s sins and their on-going punishmentwere socially apparent.The practice of individual, private confession to a clericgradually replaced penance until the Fourth Lateran Council (1215), when itwas made mandatory for all. Later, the Protestants identified compulsory con-fession to lascivious clerics as a practice that increased rather than decreasedgeneral sexual excitation. In the reformed Church confession would have noplace. Each individual would be responsible for his or her own actions in theface of God. This was not an easy option, but rather the beginning of an in-worldly asceticism. In Weber’s famous formulation, asceticism ‘now … strodeinto the market-place of life, slammed the door of the monastery behind it,and undertook to penetrate just that daily routine of life with its methodi-calness, to fashion it into a life in the world, but neither of nor for this world’(1991 [1904-5]: 154).

Just as the new order of the Protestant and Catholic Reformations gotunderway in the sixteenth century, so, too, did the witch-hunts. The judicialsystem became a means to contravene the new space of private consciencethat the reformers had begun to stake out. Officials asserted greater powerthan ever to interrogate individuals about their inner thoughts, convictions,and fantasies. These witch trials frequently involved accusations that men andwomen attended sabbaths at which they had sex with the Devil. The witch-hunting manuals developed an elaborate picture of incubi that attacked womenand succubi that copulated with men. According to the Malleus maleficarum(Kramer & Sprenger 1970 [1486]: 41 ff.), such erotic episodes occurred morefrequently to women since they were more feeble, credulous, and less self-controlled than men.

These various developments continue the story of erotic dreams and self-control begun in antiquity, a contention that emerges more clearly if weclosely consider the tenth-century Canon episcopi (Lea 1939: 38; Russell 1972:292).This text urged priests to eradicate demonic sorcery from their parishes.It also alerted clerics that some women, ‘seduced by illusions and phantasmsof demons (daemonum illusionibus et phantasmatibus seductae), believe themselves,in the hours of night, to ride upon certain beasts with Diana, the goddess ofthe pagans’ (Lea 1939: 178). In such cases, priests were instructed to teach thatthese beliefs were false delusions of the Devil. The Canon episcopi concededthat women did undergo demonic molestation but only ‘in their spirits’ (cumsolus eius spiritus). The problem, from the Church’s point of view, was the exuberant folk credulity aroused by these tales, and the laity’s apparent inability to distinguish imagined from real experiences. Thus the Canon epis-copi emphasized that

[w]hile the spirit alone endures this [demonic manipulation], the faithless mind thinksthese things happen not in the spirit, but in the body. Who is there that is not led outof himself in dreams and nocturnal vision, and sees much when sleeping that he hasnever seen when waking? Who is so stupid and foolish as to think that all these thingswhich are only done in spirit happen in the body … (Lea 1939: 179).

Uncertainty over ‘the imaginal’17 thus lay at the centre of European witchcraft.Renaissance theologians had to decide whether witches’ transformations,

flights, and sabbaths were merely dreams, and if so, whether the individuals

292 CHARLES STEWART

involved none the less merited prosecution for believing them. The issuesbegin to look very much like those posed by the desert Fathers. The differ-ence between the first ascetics and the laity during the witch craze was thatearlier a (male) individual had largely been left to monitor his own spiritualfailures. Later (male) clerics decided this matter on behalf of (female) indi-viduals. Torture and capital punishment replaced internally imposed humilityand renewed ascetic effort as responses to erotic dreams.

The matter of the reality of witchcraft, and the responsibility for dreamvisions, was never uniformly decided throughout the main period of witch-hunting, that is, in the sixteenth and seventeenth centuries. Carlo Ginzburg(1983) reveals how the authorities resolutely ignored statements by Friulianvillagers that they fought demons ‘in the spirit’, while their bodies were athome, asleep. The accused called themselves benandanti (good-doers) and imagined that their practices were fundamentally Christian. Under the duressof long interrogation, however, the benandanti changed their stories and con-fessed that they had consorted with demons ‘for real’.

The benandanti told mainly of fighting against malevolent forces in orderto safeguard the community’s harvest, and it is possible that many ‘witches’stories’ were, likewise, not particularly sexual. The inquiring authorities,however, assumed that witchcraft must involve sexual acts with the Devil andthus they pushed the stories in that direction through questioning. Judgesshowed a particular interest in the issue of whether the intercourse with thedevil was voluntary or forced, frightening or pleasurable (Kramer & Sprenger1970 [1486]: 114; Lancre 1982 [1613]: 200-1). Whether or not actual eroticnightmares or erotic dreams had occurred to the accused, there was a likeli-hood that erotic nightmare scenarios would occupy a conspicuous place inthe final confession.

Freud once rhetorically asked, ‘Why are [the witches’] confessions undertorture so like the communications made by my patients in psychic treat-ment?’ (Ginzburg 1990: 150; Roper 1994: 245). The answer would seem tolie in the shared conviction in the importance of an underlying libidinalimpulse.This sexual Urszene could be uncovered through confession, althoughboth psychoanalysts and inquisitors faced a besetting uncertainty as to whetherthese received confessions were truth or fantasy (Ginzburg 1990: 151).

In this section I have retrained attention on the persistent factor of dreams,particularly demonically distorted dreams ( phantasmata), at the heart of theEuropean witchcraft phenomenon. My contention is that dreams, eroticdreams, nightmares, and erotic nightmares all occasionally figured in witch-craft cases. The effect of the threatening manuals for prosecutors, and of the prosecutions themselves, was to funnel even innocuous dreams into anerotic nightmare formulation, thereby further defining and maximally diffus-ing a category of experience that first arose in the context of early Christianasceticism.

Philosophers and doctors

After a series of dream visions in 1619 convinced him to pursue philosophyas a vocation, Descartes returned to dreams to illustrate the proof of existencein his Meditations on first philosophy (1984 [1641]). In the first two Meditations

CHARLES STEWART 293

he posed the radical question: ‘What if I do not exist at all, but am, rather,the figment of a dream?’ He replied that, whereas in a dream he could notpossess awareness that he was thinking, in a normal state of consciousness hecould possess such an awareness. Meta-cognition thus established the groundsof reality as opposed to fantasy. Hobbes also assumed this distinction in con-tending that demons were ‘but idols or Phantasmes of the brain, without realnature of their own distinct from human fancy’ (Hobbes 1957 [1651]: 398;chap. 44; see also 423, 432; chap. 45). Although this view had been expressedalready in the Canon episcopi, it increasingly became a fundamental supposi-tion in science and philosophy.

Medical doctors also produced numerous exclusively physical explanationsfor dream phenomena. Beginning with Aristotle, melancholics had neverceased to be prime candidates for disturbed dreams. Now sceptical doctorslike Weyer (1991 [1583]: 232) attributed incubus experiences to ‘phlegm andmelancholia’, while Burton linked it to eating ‘black meat’ such as hare andvenison in his Anatomy of melancholy (1927 [1620]: 190). In one case, a woman’sclaim to have borne the devil’s baby was medically diagnosed as a case ofchronic constipation and its final relief (Lea 1939: 1470).

Between 1650 and 1850 no fewer than twenty-five tracts were writtenabout the nightmare by doctors and scholars ( Jones 1971 [1931]: 14). In thisperiod, the term ‘incubus’ went from denoting an independent demonic beingto denoting an objective set of physical conditions. Certainly the word ‘night-mare’ also began to lose vital reference to the supernatural ‘mare’ spirit con-tained in its etymology. Physical and medical explanations naturalized both theerotic dream and the nightmare, although there was still sporadic support forthe idea that external spirits produced dreams.18

The amalgamation of erotic dreams and nightmares which had begun inthe first centuries CE now started to come apart. Medical tracts from thepost-witchcraft period considered the nightmare to be characterized primar-ily by dread and a feeling of suffocation. Erotic sensations might accompanythis, but they were marginal. A Royal Navy doctor named Waller mentioned‘priapism’ as a symptom of the nightmare in a very brief Latin passage, as ifthe subject were embarrassing (1816: 25). His assertion that males suffer night-mares more frequently than females indicated the appropriation of ‘incubus’as a nosological category unfettered by Church misogyny. Like Waller,MacNish (1830: 73, 124 ff.) also focused on the feature of sleep paralysis ascharacteristic of the nightmare and speculated on asthma, angina, and indi-gestion as possible causes. Erotic dreams came to be placed in a separate cat-egory centring on masturbation, the subject of two treatises in 1760 ( Jaccard1975: 11). Erotic dreams involved a form of mental masturbation, delectatiomorosa, and became a subject for the emergent discipline of sexology (Ellis1936 [1898]; Kinsey, Pomeroy & Martin 1948).

This is not to suggest that people no longer experienced erotic nightmares.Just like the ancient Greeks in the period preceding the formulation of theerotic nightmare complex, people in the Enlightenment period did, occa-sionally, feel ‘Desire with loathing strangely mixed’.19 Without the buttress ofofficial belief in witchcraft, however, the complex began to disintegrate intotwo separate experiences. Without a background of shared cultural concepts,and lacking a concise vocabulary to express the synthetic combination, the

294 CHARLES STEWART

erotic nightmare experience entered into a phase of ‘hypocognition’ (Levy1984: 227). Levy’s term enables us to avoid the Whorfian assumption that lackof a lexeme indicates a corresponding lack of appreciation or perception of agiven object. The basic sensations continued to be subjectively felt but couldnot easily be expressed.

Undoubtedly there was now an element of repression (in the psychoana-lytic sense) behind the hypocognition of the erotic nightmare. Its discursivedisappearance reflected an unwillingness, and ultimately an inability, to bringthe phenomenon to conscious awareness. This tense, repressive silence wasqualitatively different from the silence around the erotic nightmare in classi-cal Greek antiquity. A millennium and a half of Christian teaching had ren-dered the erotic nightmare problematic, even though science had momentarilydissolved it.

In the Victorian period, public morality increasingly suppressed the repre-sentation of the erotic. The Obscene Publications Act of 1857 indexes thissense of public decorum, as does the coinage of the term ‘pornography’ fromvenerable ancient Greek roots.20 Indeed, erotic artworks from the ancientworld, particularly the recently excavated frescoes from Pompei, were amongthe factors provoking these public reactions. The offending penchants of theancients were accommodated by the foundation of secret museums such asthe British Museum’s ‘Secretum’, established in 1865, or the ‘Gabinetto degliosceni’ created earlier at the Museo Borbonico in Naples (Kendrick 1987:11). Erotic dreams were similarly relegated to their own secretum.

Malinowski’s erotic nightmare

Malinowski’s diaries, kept during his fieldwork between 1914 and 1918, andnever meant for publication, afford a rare perspective on the male manage-ment of desire during the immediately post-Victorian period. By day heresisted the attractive local girls, while under the mosquito netting at nighthe recollected past relationships with women in Europe and Australia.21 Thediary chronicles the near-penitential burden which he undertook in order torealize his ambitions as an anthropologist. Malinowski coped with thoughtsthat he variously terms ‘impure’, ‘lecherous’, ‘sensual’, or ‘erotic’ and recon-ciled himself to the ‘metaphysical regret’ that ‘You’ll never fuck them all’ (1967:114; Torgovnick 1990: 227). He followed a regime of Swedish gymnastics,dosed himself with arsenic and quinine, and swore off reading novels, realiz-ing that

[p]urity in deeds depends on purity of thought, and I resolve to watch myself right downto the deepest instincts … I can repress occasional violent whoring impulses by realizingthat it would get me nowhere, that even if I possessed women under these conditions,I would merely be sloshing in the mud. The most important thing is to have a strongaversion for sloshing in the mud (onanism, whoring, etc.). And to seek out everythingthat builds up such an aversion (1967: 181).

Malinowski, of course, established extended fieldwork as a rite of passageon the way to achieving full status as a social anthropologist. Such fieldworkinvolves combating the same deadly sins such as despondency, self-regard, and

CHARLES STEWART 295

sensuality which had been identified by the early Christian ascetics.The moveoff the verandah might just as well have been a move into the desert.22 Anthro-pologists who could resist the demons of temptation emerged transfigured asprofessionals; they achieved an ‘ultimate mastery of things’ (Malinowski 1967:175).

Malinowski had abandoned Catholicism at an early age. His asceticism didnot grow out of any active religious faith on his part, although it can, perhaps,be traced back via Weber’s in-worldly asceticism to early Christian monasti-cism. A stronger case can be made for regarding psychoanalysis as the con-tinuation of religious confession in a deconsecrated, medical format.Psychoanalysis adapted the formalized Catholic rite and applied it to the deep-ened idea of inner conscience that Protestantism had fostered (Foucault 1978:68; Webster 1995: 350).The analyst prompted the patient to search his or hermemory for pathogenic secrets the disclosure of which led to cathartichealing.A secular therapy, psychoanalysis fitted the needs of a secularizing edu-cated middle class. When Malinowski first read Freud as a teenager it isreported that ‘he felt that he was a complete case of the Oedipus situation’(Wayne 1985: 532). The following dream, recorded on 8 June 1918, near theend of his last stint of fieldwork, further reveals the degree to which he hadinternalized Freud’s categories:

This morning I woke early (I did not sleep very well and had two horrid … dreams). In the firstone, which was of a Freudian type, feeling of sinfulness, evil, something loathsome, com-bined with lust – repulsive and frightening. What does it come from? And this feelingof wickedness, which rises to the surface (1967: 290, ellipsis indicates ‘extremely intimateobservations’ omitted by the editor).

This dream occurred just before the news of his mother’s death reached himin the Trobriands, and just as he was deciding definitively to drop Nina Stirling (NS) and become engaged to Elsie Masson (ERM). Earlier he explic-itly (1967: 245) associated his love for ERM with ‘feelings of child for mother(vide Freud’s theory)’, and he had dreamt that his mother reproached him fornot marrying NS (1967: 202). These two clues might support an interpreta-tion of this dream along the orthodox psychoanalytic lines laid out by Jones– an erotic nightmare produced by the repression of incest desires. Malinowskihimself accepted it as a dream of the ‘Freudian type’, but I think an alterna-tive interpretation is possible.

Whereas Jones would emphasize unconscious phantasy and its repression inthe production of this nightmare, I would argue that it resulted more straight-forwardly from conscious fantasy and its suppression.23 It is continuous withMalinowski’s sensuous daytime thoughts, his commitment to sexual abstinenceout of loyalty to his fiancée(s), and his repeated self-excoriations for havingsuch thoughts at all. His erotic nightmare is the result of his own auto-terrorization in the cause of ambitious self-making. The power of this self-making impulsion is apparent already in his next diary entry, recorded sometwo weeks later. Malinowski reports that he is hard at work, ‘almost indiffer-ent in relation to grief ’. Furthermore: ‘external ambitions keep crawling overme like lice. F[ellow of the]. R[oyal]. S[ociety]. – C[ompanion of the]. S[tarof ]. I[ndia]. – Sir’. He also contemplates his entry in Who’s who (1967: 291).

296 CHARLES STEWART

Self-making or repression?

Dreams appear to be fundamentally involuntary events – they just occur. Likethe death of a parent, or first orgasm, they expose the individual to powerfulnovelties that must somehow be accommodated in the psyche and per-sonality. The psychoanalysts Abraham and Torok (1994) have termed this transformation of the self in the face of striking internal and external events‘introjection’.Their usage departs from the traditional psychoanalytic sense ofthis term as the reinternalization of something that was initially within thepsyche but subsequently projected outside it (Crapanzano 1977: 12). Abrahamand Torok relativize repression as but one form of disturbance in the processof introjection. They emphasize, as I also contend, that these disturbances do not necessarily arise from childhood experiences, or from inheritedinstincts.They may result directly from current individual life situations (Rand1994: 11).

Malinowski’s diaries, like the comprehensively analysed inner desires of the early monastics, reveal conscious strategies for dealing with consciouslyperceived desires and images. Signs of sexuality are suppressed but, unlikerepression, this is a voluntary privation on the way to what is perceived as amore important undertaking – the making of the self according to a chosen‘aesthetic’ (Foucault 1985: 253). All self-making involves selective suppressionand this volition can, evidently, carry over into dreams. Just as the everydaythoughts and actions of Plato’s democrat made for calm dreams, so Malinowski’s cultivation of ‘pure thoughts’ was also a means of preventingunwanted lewd eruptions in fantasies or dreams. His strategy was entirely con-sistent with the teachings of the high Victorian doctor William Acton that ‘if [a] man has allowed his thoughts during the day to rest upon libidinoussubjects, he finds his mind at night full of lascivious dreams’ (Marcus 1966:24). In such cases the principles of one’s waking life personality are appliedduring sleep; the action of suppression carries over into dreams.

The history that I have presented above uncovers a perennial suppositionthat the phenomenological dream could be altered by force of will and self-preparation. Christian ascetic writers, for example, explicitly extended Stoictechniques for judging waking perceptions to the monitoring of one’s reac-tions to images in sleep. Bedtime prayers against demonic dreams presentedanother attempt to suffuse sleep with the force of will. And the penitential ofCummean, which assigned different penances to someone who ‘desires to sinduring sleep’ and someone who ‘is willingly polluted during sleep’ (Bieler1963: 7), evidently assumed that dream contents were subject to volition.

Aristotle, as we saw, considered consciousness during dreams a rarity and,until recently, most dream researchers shared this view. Standard definitions of the dream (Hadfield 1954: 17; Rycroft 1981: 7) followed Descartes in considering it to occur independently of, and uncontaminated by, consciouswill. Lucid, or conscious, dreaming, where people achieve consciousness in thedream, has thus usually been explained as a phenomenon of waking(Rechtschaffen 1978: 100).

In the last twenty years, this consensus has been shaken not only by dreamlaboratory demonstrations of lucidity (LaBerge 1985), but also by the apparent success of manuals and workshops that teach people how to gain

CHARLES STEWART 297

consciousness in their dreams (e.g. LaBerge & Rheingold 1990). Evidently,then, dreams can be subject to intentional designs, even at the deep level oftheir first realization. The techniques for achieving lucidity are neither mys-terious nor excessively demanding. Many report their first lucid dream experi-ences after merely learning about the idea (Green & McCreery 1994: 114).

The effect of lucid dreaming in the service of achieving a vision of self offersa plausible mechanism by which erotic dreams could go from non-nightmaresto nightmares and back again over the last two and a half millennia. In Theinterpretation of dreams, Freud (1953 [1900]: 572) remarked that one good featureof conscious dreaming was that it allowed one to alter disturbing dreams whilethey were still in process. If a dream led one into a sexually stimulating situ-ation, he suggested, the dreamer ‘can think to himself, “I won’t go on withthis dream any further and exhaust myself with an emission; I’ll hold back fora real situation instead.” ’ A negative perception of erotic dream thoughts thusleads to their foreclosure. In the ascetic ‘anthropology’ of the early Church,erotic ‘thoughts’ (logismoi) were held to be demon-inspired. Excluding themthus often took the form of a dramatic battle against demons. The juxtaposi-tion of conscious will and non-consciously controlled desires in these dreamsfurnishes yet another reason for considering them ‘mixed’ dreams.

Of course, the same initial erotic dream could take a completely differentturn depending on the current self-making project of the person who has it.Patricia Garfield, the author of books entitled Creative dreaming and Pathwayto ecstasy (1974; 1979), reported that about two-thirds of her lucid dreams hadsexual content, and about half culminated in orgasm. Her intentional cultiva-tion of these climactic conclusions represents a project of self-fulfilmentexactly the converse of those encouraged by early ascetic practice, Victorianprecepts, or the immediate post-Victorian inner worlds of Malinowski andFreud.

One further illustration of the unproblematic cultivation of erotic dreamsis the manufacture and use of statues representing ‘otherworld mates’ by theBaule of Côte d’Ivoire (Ravenhill 1996). Both men and women can com-mission the carving of these small statues and a night of the week is set asidewhen one sleeps alone in order to have dream encounters, including eroticexperiences, with these idealized otherworld beings.The bodies of the femalesare accentuated, while the clothing, indicating status, of the male statues ismore emphasized (Figs. 3, 4). This reveals the fantasy features most appealingto male and female dreamers respectively. These dream experiences with otherworld mates apparently serve as a complement to, and at the same timea release from, one’s real life partner and the normal social world. The Bauleexample provides one last illustration of how dreams may be continuous with,and expand upon, current waking fantasies of the self. This is a dimensionthat reliance on the psychoanalytic term ‘repression’, which emphasizes dreamsas distorted products of a discontinuous unconscious, might cause one to miss.

Conclusion: viral history

From the rise of Christian asceticism until the Enlightenment, the eroticnightmare complex, the incubus to give it its salient indigenous term, lay upon

298 CHARLES STEWART

the brain of Western societies. Naked evidence of one’s own lust posed aproblem and the introjection of erotic dream experiences was consequentlyfraught. The popular cultivation of orgasmic dreams today would seem tosuggest the end of the erotic nightmare, but the matter is not so simple.Recent developments suggest that the erotic nightmare might be headed fora more mainstream revival. In order to account for this I must take one briefstep backward before going forward.

CHARLES STEWART 299

Figure 3. Statue of an otherworld woman (c. 1920). Courtesy of the Fowler Museum of Cul-tural History, UCLA (Ravenhill 1996: 49).

At the moment when science and philosophy had almost completelysubdued the animate, demonic nightmare, the romantics revitalized it in a sortof backlash against the tyranny of reason. On holiday in Switzerland, a selectcompany that included Lord Byron, his personal physician John Polidori, PercyShelley, and his soon-to-be wife Mary Wollstonecraft Godwin regularly gath-ered around the fireplace. Confinement by unseasonably cold weather andperpetual rainstorms brought boredom, which they countered by reading

300 CHARLES STEWART

Figure 4. Statue of an otherworld man (c. 1965). Courtesy of the Fowler Museum of Cultural History, UCLA (Ravenhill 1996: 47).

aloud from a volume of German folktales in French translation. They thenturned their energies to composing their own horror stories. Byron’s taleabout a vampire (‘A fragment’, 1816), Polidori’s appropriation of it (Thevampyre, 1819) and Mary Shelley’s Frankenstein (1818) all arose from thiscontest. Only eighteen years old at the time, Mary Shelley could not thinkup an idea until one night she had a waking dream, which she described inher introduction to the 1831 reissue of Frankenstein (Shelley 1993: 196). Inthis dream she saw Dr Frankenstein bringing his monster to life. Anotherfamous horror story of the nineteenth century, Bram Stoker’s Dracula, alsotook form in a dream. Stoker’s diaries reveal that his idea for this novel derivedfrom an actual nightmare in which he envisioned a girl trying to kiss him,not on the lips, but on the neck (Frayling 1996: 68).

Conceptions of the erotic nightmare gained power and unity in the genreof gothic horror fiction and through Fuseli’s famous painting, which also drewon his personal dream experiences.24 In time these works of art and fictionwere transmuted into twentieth-century films with an even wider diffusion.Horror stories and films preserve precisely the spellbinding, terrifying ‘grip’often felt in the erotic nightmare (Todorov 1973: 25, 32). Gothic horror thusre-poses the problem of the imaginal from the opposite direction of thedream. In it the fictional converts into a personally felt emotional experience,while in the case of nightmares one struggles to classify a personal emotionalexperience as fiction (‘It was only a dream’). The felt phenomenology of themedieval incubus thus evidently jumped from the category of personal dreaminto the realm of fiction just as it began to lose credence as a specific kindof dream experience in educated circles.

The process of dismembering the erotic nightmare seems to have beenlargely completed by the late twentieth century.The psychiatrist Mack (1970:vii, 224) doubted that nightmares had any direct relation to eroticism and herejected Jones’s repression of incest aetiology as too narrow (Hufford 1982:130). A large sample of nightmares collected in the Boston area did not revealany sexual content (Hartmann 1984: 61).The term incubus itself was retainedby Broughton (1970) to refer to night terrors occurring in NREM deep sleep,but the broad public seems to have grown completely unfamiliar with theterm ( Jarcho 1980: 253). In 1996 Reebok launched a women’s training-shoenamed ‘Incubus’ and it was a full year, and over 50,000 unit sales later, beforeanyone pointed out that the incubus was a demon that molested women intheir sleep.

It appears, however, that having migrated into fictional genres, many of the elements of the erotic nightmare have now begun to re-enter the domainof non-fictional human experiences in accounts of alien abduction. Whatbegan in the 1940s as periodic sightings of ‘flying saucers’ has developed intoan increasing number of more detailed reports of closer encounters withaliens. The very first alien abduction incident gives an idea of this emergentcategory of experience. In 1957 a twenty-three-year-old Brazilian named Antonio Villas-Boas was forcibly taken aboard a UFO by five aliens. There,according to his account (Matheson 1998: 40 ff.), he was drenched in a strangeviscous liquid and a blood sample was taken from his chin before he hadsexual intercourse with one of the aliens. He described her as attractive but emitting grunting noises that left him with a ‘disagreeable impression’,

CHARLES STEWART 301

as if he had been with an ‘animal’ (Matheson 1998: 43). Many further abduc-tions have since been reported and they often begin with the victims in bed, perhaps asleep, when a feeling of paralysis comes over them (Bryan 1996:18).

Numerous abductees believe that they might be dreaming the events, orthat they occur in an intermediate phase between fantasy and normal con-sciousness. The researchers often discount these opinions and treat the eventsas real, much as the inquisitors did during witch accusations.These abductionaccounts are very various, but they frequently involve a feeling of dread, terror,and paralysis. Clinical invasions of one’s reproductive system are frequentlyaccompanied by mixed feelings of pleasure and horror. Women, who consti-tute the majority of abductees, report that the aliens perform something called‘mindscan’ on them, which produces feelings of love for the aliens and thewish to submit to them. Sometimes the aliens induce rapid sexual arousal,even orgasm, although victims report this as an unpleasant experience ( Jacobs1992: 106).

The example of alien abductions carries my account up to the present,and probably some distance into the future. It dispels any urge to considerthe history of erotic dreams as a progressive mastering of the frighteningimagery modulating the experience of sexual desire. Although alien abduc-tion accounts resemble the Genesis 6 story of the union of mortal womenwith the sons of God (also known as the ‘Watchers’), or the operation ofmedieval incubi and succubi, this history does not document a steady continu-ity for the erotic nightmare. The dream, or nightmare, presents an attractivemodel for historical process because it conveys the idea of compulsion.One is borne along, gently in a dream, brutally in a nightmare, just as one is carried along in the stream of historical events, unable to step outside it.Yet this image does not apply in this case precisely because of the rupturesin the discursive appearance of the erotic nightmare. Nor does it seem satis-factory to label its periodic re-emergence as ‘cyclical’.This image of mechani-cal regularity is too thin and unilluminating. What needs to be captured inthe case of erotic nightmares is the manner in which a particular bundle ofemotions and sensations has been preserved even through transformations ofgenre.

The erotic nightmare has mutated and ‘wandered’ because, as Hacking sug-gests (1995b), social classifications of human experience are bound to stimu-late reactions in the people so classified. They may embrace the descriptionof their experience, or hyper-conform to it; but even if they reject it, theyoften subsequently reinvent it with a new name (Hacking 1995b: 374). Themigration and mutation of the erotic nightmare across genres resembles thehistory of a virus that passes back and forth between species, even under-going long periods of latency, before bursting out again. Mary Shelley cap-tured this process of infection when, awakening from her dream, she imme-diately realized its potential for the ghost story she had been trying tocompose. ‘ “I have found it! What terrified me will terrify others; and I needonly describe the spectre which had haunted my midnight pillow” ’ (1993:196). Precisely this sort of transmission across social contexts (for instance,from medieval folk-story to witch prosecutor’s manual, to gothic novel or psychoanalytic theory) seems to have ensured the erotic nightmare’s survival.

302 CHARLES STEWART

As Lévi-Strauss (1963: 217) asserted in the case of retellings of myths, sub-sequent narrations of the erotic nightmare (no matter how analytic) becomefurther bearers of it.

The idea of a viral history is purely analogical; the incubus may be rhetori-cally and physically contagious, but it is not a pathogen like HIV. This studyof the erotic nightmare complex does, however, supplement Sperber’s (1985)‘epidemiology of representations’ by calling attention to the role of feelingsand emotions in the spread of ideas.The diachronic transmission of the eroticnightmare depended perhaps more on the transfer of a powerful bundle ofaffect than on the reception of cognitive ideas. Representations of the eroticnightmare have indeed transformed as they went from individual minds tocollectively available media and back again, as Sperber envisaged (1985: 75).Distortions in transmission do not, however, account for the full dynamic ofthe erotic nightmare. Clearly the various internal rules of specific domains ofthought (asceticism, medicine, philosophy, fiction, psychiatry) acted as cruciblesof cultural historical change. Early Christian ascetics sought to eliminate eroticdreams as part of a quest for spiritual perfection. Renaissance inquisitors multi-plied the number of erotic nightmare scenarios by flushing out, or imputing,such dreams in order to secure convictions. Post-Enlightenment doctors brokeerotic dreams and nightmares into discrete constituent units in order to treatthem therapeutically. And the stories of present-day abductees proliferate andgain added complexity as they meet with enormous public interest and occa-sional monetary reward. Transfer from one to another social context has thussubjected the erotic nightmare to radical yet highly unco-ordinated concep-tual escalations or contractions. This account alerts us that the erotic night-mare may currently still exist, and also that an historical approach is necessaryin order to locate it.

NOTES

I thank the Anthropology Department of the London School of Economics for inviting me to present the Malinowski Memorial Lecture for 1998. As I acknowledged on that occa-sion, I am extremely grateful to John Campbell, Margaret Alexiou and Adam Kuper for theirinspiration over many years. In preparing this article I have received valuable advice and assis-tance from: Margaret Alexiou, Christina Angelidi, George Calofonos, Diskin Clay, Susan GuettelCole,W. Robert Connor, James Davidson, Jason Davies, Dimitris Dikeos, Mary Douglas, DyanElliott, Alida Gersie, Alan Griffiths, Roland Littlewood, Daniel Miller, Leonard Muellner,Rodney Needham, Deena Newman, Buck Schieffelin, Richard Sorabji, Rebecca Spang, PaulStrohm, Bradley Vaughn, Chris Waters and James Woodburn. I am particularly indebted to PeterBurian for calling my attention to the interest of alien abductions. Of course none of the aboveare to be held accountable for the views I present here. This research was aided by a Fellow-ship at the National Humanities Center and a grant from the National Endowment for theHumanities.

Passages from classical and patristic authors are cited by internal textual division, so that thepassage may be located in any edition or translation of the work.

1 Foucault had long planned a fourth volume in his History of sexuality. Provisionally titledThe confessions of the flesh, it would have dealt centrally with early Christian asceticism.

2 Johnson (1998) represents an exception.3 Freud already distinguished conscious suppression (Unterdrückung) from unconscious repres-

sion (Verdrängung) in The interpretation of dreams (1953 [1900]: 606-7; Laplanche & Pontalis 1973:438).

CHARLES STEWART 303

4 For an overview of these and other ‘culture-bound syndromes’, see Simons and Hughes(1985), who also include sleep paralysis (the classic nightmare) as one of the syndromes.

5 Freud had a reproduction of Fuseli’s The nightmare hanging in his consulting room atBerggasse 19 (Frayling 1996: 10). The picture was a gift from Ernest Jones, who used it as thefrontispiece for his book On the nightmare (1971 [1931]). For an excellent study of the classicnightmare, see Hufford (1982).

6 In this respect dreamers are comparable to the mentally ill, who mistake hallucinations ordelusions for reality. Aristotle explicitly makes this comparison (On dreams, 460b15) and laterphilosophers also regularly contrasted dreaming with madness (Pigeaud 1983).

7 Erotic dreams were tests of what Foucault termed ‘governmentality’ – the ethical basis ofthe right to govern (1997). The life of Gandhi furnishes a clear illustration. In his final yearsGandhi slept in the same bed as young women, claiming to be testing his ability not to becomearoused. He was quoted as saying, ‘If I can master this, I can still beat Jinnah’ (Erikson 1970:404; Alter 2000).

8 In Paulus Aeginetus page 30b, cited in Roscher (1900: 18, 112-13).9 I thank Jeremy Tanner for his discussion of this relief.10 On acute diseases and On chronic diseases are attributed to the fifth-century CE Latin writer,

Caelius Aurelianus.These texts are, however, believed to be Latin translations of Greek treatisesby the second-century CE physician, Soranus.

11 Seneca (first century CE) later modified the Stoic position to hold that there were certain‘first movements’, such as shuddering when splashed with cold water, or experiencing sexualarousal (On anger, 2.1-3), that could never be subject to mental control and thus were not pas-sions, but just a physical ‘impulse of the body’ (corporis pulsus). I thank Richard Sorabji for hisobservations drawn upon here (see also Sorabji 1997: 200).

12 Greek magical papyri XVIIa; in Betz (1986: 253). For more on demons sending (erotic)dreams, see Eitrem (1991) and Faraone (1999).

13 Found in the first-century CE author, Aëtius, Placita, 5.2.3; text and translation in vonStaden (1989: 386).

14 Galen considered erotic dreams as textbook examples of the category of dreams thatreflected an individual’s physical state: ‘men full of sperm will imagine that they are havingsexual intercourse’ (On diagnosis from dreams, in Oberhelman 1983: 46).

15 The sin of accidie (boredom, despondency) – also known as the noonday demon – provides one example of a demonic thought that allied the irascible and sensuous parts of thesoul and ‘suffocated the intellect’ (Evagrius, Praktikos, 36). Evagrius considered accidie ‘the heavi-est of the demons’ (Praktikos, 12). I thank George Calofonos for his discussion of these ideas.

16 An indication of the currency of the term may be found in Augustine’s City of God, 15.23.17 My use of this term perhaps differs somewhat from its use in Jungian circles and else-

where (cf. Tedlock 1987: 3). I use ‘imaginal’ to refer to a state of consciousness in which onehas the impression that what one is witnessing is absolutely ‘real’ and independent of one’smind, although one is, in fact, only imagining it. The term ‘imaginary’, by contrast, implies anawareness, even in the moment of imagining, that what one beholds is only a product of one’simagination. Dreams, visions, hallucinations, and apparitions are, generally, experienced imagi-nally and then subsequently accounted for as imaginary.

18 Andrew Baxter’s Enquiry into the nature of the human soul (1737) attributed frighteningdreams to external spirits because, he reasoned, the soul would not frighten itself (Ford 1999:178).

19 Coleridge, ‘The pains of sleep’ (1816). In Coleridge (1912: 390).20 ‘Pornography’ appears for the first time in Webster’s Dictionary of 1864 (Kendrick 1987:

13).21 Malinowski’s daytime erotic thoughts often concerned the local women but, as Tedlock

has pointed out (1994: 289), these women never figured in the dreams he recorded.22 Unlike Christian ascetics, however, anthropologists receive little preparation for the psycho-

sexual travails of fieldwork. In his Diary Malinowski thus inadvertently broached importantissues that have only recently begun to be discussed within the profession (Kulick & Willson1995; Markowitz & Ashkenazi 1999; Wengle 1988).

23 Susan Isaacs distinguished ‘fantasy’ (conscious daydreams, fictions, etc.) from ‘phantasy’(the primary content of unconscious mental processes). See Hook (1979; 1994). Laplanche and

304 CHARLES STEWART

Pontalis (1968: 11) disputed Isaacs’s distinction, although elsewhere they make it clear that Freudconceived phantasy, and primal phantasy, as possibly archetypal, phylogenetic inheritances(Laplanche & Pontalis 1973: 315, 332).

24 According to Fuseli: ‘One of the most unexplored regions of art are dreams, and whatmay be called the personification of sentiment’ (Tomory 1972: 181). He painted The nightmarefresh with the disappointment of having been turned down in his proposal for the hand ofAnna Landolt. On the obverse of the original canvas Fuseli had begun to paint the portrait ofa young woman, possibly Anna (Powell 1972: 60). Fuseli’s obsession with Anna is made overinto her erotic nightmare desire for him (as incubus).

REFERENCES

Abraham, N. & M. Torok 1994. The shell and the kernel: renewals of psychoanalysis (ed. and trans.N. Rand). Chicago: University Press.

Aeschylus, Libation bearers 1983. In Aeschylus, 1: Oresteia (ed. D. Grene & R. Lattimore). Chicago:University Press.

Alter J. 2000. Gandhi’s body: sex, diet and the politics of nationalism. Philadelphia: University ofPennsylvania Press.

Angelidi, C. forthcoming.Aisthiseis, sexoualikotita kai optasies. In Anokhi kai katastoli stous mesouskhronous (ed.) K. Nikolaou. Athens: Institute for Byzantine Research/NHRF.

Aristotle, On dreams 1990. In Aristotle on sleep and dreams, D. Gallop, Peterborough, Ontario:Broadview Press.

Artemidorus, Interpretation of dreams. 1990. The interpretation of dreams: oneirocritica (trans. R.J.White), Torrance, Calif.: Original Books.

Asad,T. 1993. Notes on body pain and truth in medieval Christian ritual. In Genealogies of reli-gion: discipline and reasons of power in Christianity and Islam, T. Asad, 83-124. Baltimore, Md.:Johns Hopkins University Press.

Augustine, City of God 1957-72. 7 vols. (Loeb Classical Library). London: Heinemann.Betz, H.D. (ed.) 1986. The Greek magical papyri in translation. Chicago: University Press.Bieler, L. (ed.) 1963. The Irish penitentials. Dublin: Institute for Advanced Studies.Boardman, J. & E. La Rocca 1978. Eros in Greece. London: J. Murray.Brakke, D. 1995. The problematization of nocturnal emissions in early Christian Syria, Egypt

and Gaul. Journal of Early Christian Studies 3, 419-60.Broughton, R. 1970. The incubus attack. In Sleep and dreaming (ed.) E. Hartmann, 188-92.

Boston: Little & Brown.Bryan, C.D.B. 1996. Close encounters of the fourth kind. London: Orion.Burton, R. 1927 (1620). The anatomy of melancholy (ed. F. Dell & P. Jordan-Smith). New York:

Farrar & Rinehart.Caelius Aurelianus, Treatise on acute diseases; Treatise on chronic diseases 1950. In Caelius Aurelianus:

on acute diseases and on chronic diseases, I.E. Drabkin. Chicago: University Press.Cassian ( John), Conferences 1955-9. In Conférences (ed. E. Pichery). 3 vols. (Sources Chrétiennes

42, 54, 64). Paris: Cerf.——— Institutions 1965. In Institutions cénobitiques (ed. J.-C. Guy). (Sources Chrétiennes 109).

Paris: Cerf.Chantraine, P. 1977. Dictionnaire étymologique de la langue grecque. Paris: Klincksieck.Climacus ( John), Ladder 1982. In John Climacus: the ladder of divine ascent (trans. C. Luibheid &

N. Russell). Mahwah, N.J.: Paulist Press.Coleridge, S.T. 1912. The complete poetical works (ed. E.H. Coleridge). Oxford: Clarendon

Press.Crapanzano, V. 1977. Introduction. In Case studies in spirit possession (eds) V. Crapanzano & V.

Garrison, 1-40. New York: Wiley.Dean-Jones, L. 1992.The politics of pleasure: female sexual appetite in the Hippocratic corpus.

Helios 19, 72-91.Descartes, R. 1984 (1641). Meditations on first philosophy. In The philosophical writings of Descartes,

vol. 2 (ed. J. Cottingham, R. Stoothoff & D. Murdoch). Cambridge: University Press.

CHARLES STEWART 305

Eitrem, S. 1991. Dreams and divination in magical ritual. In Magika hiera (eds) C. Faraone &D. Obbink, 175-87. Oxford: University Press.

Elliott, D. 1999. Fallen bodies: pollution, sexuality and demonology. Philadelphia: University of Pennsylvania Press.

Ellis, H. 1936 (1898). Studies in the psychology of sex, vol. 1, pt 1. (Revised edition). New York:Random House.

Erikson, E. 1970. Gandhi’s truth. London: Faber.Evagrius of Pontus, Praktikos 1971. In Évagre le poutique: traité pratique, ou le moine, vol. 2 (ed.)

A. Guillaumont & C. Guillaumont. (Sources Chrétiennes 171). Paris: Cerf.Faraone, C. 1999. Ancient Greek love magic, Cambridge, Mass.: Harvard University Press.Fögen, M.T. 1998. Unto the pure all things are pure: the Byzantine canonist Zonaras on noc-

turnal pollution. In Obscenity: social control and artistic creation in the European Middle Ages (ed.)J. Ziolkowski, 260-78. Leiden: Brill.

Ford, J. 1999. Samuel Taylor Coleridge and the pains of sleep. History Workshop Journal 48,169-86.

Foucault, M. 1978. The history of sexuality, vol. 1: An introduction. New York: Pantheon.——— 1985. The history of sexuality, vol. 2: The use of pleasure. New York: Pantheon.——— 1986. The history of sexuality, vol. 3: The care of the self. New York: Pantheon.——— 1997. Subjectivity and truth. In The essential works of Michel Foucault, 1954-1984, vol.

1: Ethics: subjectivity and truth (ed. P. Rabinow), 87–92. New York: New Press.Frayling, C. 1996. Nightmare: the birth of horror. London: BBC Books.Freud, S. 1953 (1900). The interpretation of dreams. (Standard Edition 4, 5). London: Hogarth.Gamillsches, E. 1969. Etymologisches Wörterbuch der französischen Sprache. Heidelberg: Carl Winter.Garfield, P. 1974. Creative dreaming. New York: Ballantine.——— 1979. Pathway to ecstasy. New York: Holt, Rinehart & Winston.Gell, A. 1977. Magic, perfume, dream. In Symbols and sentiments (ed.) I. Lewis, 25-38. London:

Academic Press.Ginzburg, C. 1983. The night battles: witchcraft and agrarian cults in the sixteenth and seventeenth

centuries. Baltimore, Md.: Johns Hopkins University Press.——— 1990. Freud, the wolf-man, and the werewolves. In Myths, emblems, clues, C. Ginzburg,

146-55. London: Hutchinson Radius.Green, C. & C. McCreery 1994. Lucid dreaming: the paradox of consciousness during sleep. London:

Routledge.Greenhouse, H. 1974. Penile erections during dreams. In The new world of dreams (eds) R.Woods

& H. Greenhouse, 296-7. New York: Macmillan.Grottanelli, C. 1999. On the mantic meaning of incestuous dreams. In Dream cultures: explo-

rations in the comparative history of dreaming (eds) D. Shulman & G. Stroumsa, 143-66. Oxford:University Press.

Guillaumont, A. 1971. Étude historique et doctrinale. In Évagre le pontique: traité pratique, ou lemoine, vol. 1 (eds A. Guillaumont & C. Guillaumont), 21–125. (Sources Chrétiennes 170).Paris: Cerf.

Hacking, I. 1995a. Rewriting the soul: multiple personality and the sciences of memory, Princeton:University Press.

——— 1995b. The looping effects of human kinds. In Causal cognition (eds) D. Sperber, D.Premack & A.J. Premack, 351-83. Oxford: University Press.

Hadfield, J.A. 1954. Dreams and nightmares. Harmondsworth: Penguin.Hartmann, E. 1984. The nightmare: the psychology and biology of terrifying dreams. New York: Basic

Books.Herodotus, History 1998. The history: herodotus (trans. D. Grene). Chicago: University

Press.Hobbes, T. 1957 (1651). Leviathan (ed. M. Oakeshott). Oxford: Blackwell.Hook, R.H. 1979. Phantasy and symbol: a psychoanalytic point of view. In Fantasy and symbol

(ed.) R.H. Hook, 267-91. London: Academic Press.——— 1994. Psychoanalysis, unconscious phantasy and interpretation. In Anthropology and

psychoanalysis (eds) S. Heald & A. Deluz, 114-27. London: Routledge.Hufford, D. 1982. The terror that comes in the night: an experience-centered study of supernatural assault

traditions. Philadelphia: University of Pennsylvania Press.

306 CHARLES STEWART

Jaccard, R. 1975. L’exil intérieure. Paris: PUF.Jacobs, D. 1992. Secret life: firsthand, documented accounts of UFO abductions. New York: Simon &

Schuster.Jarcho, S. 1980. Some lost, obsolete, or discontinued diseases: serous apoplexy, incubus, and

retrocedent ailments. Transactions and Studies of the College of Physicians of Philadelphia, pt 2,ser. 5, 241-66.

Johnson, A. 1998. Repression: a reexamination of the concept as applied to folktales. Ethos 26,295-313.

Jones, E. 1971 (1931). On the nightmare. New York: Liveright.——— 1974 (1932). Beliefs concerning the nightmare. In Psycho-myth, psycho-history, E. Jones,

110-13. New York: Hillstone.Kendrick, W. 1987. The secret museum: pornography in modern culture. New York:Viking.Kinsey, A.C., W.B. Pomeroy & C.E. Martin 1948. Sexual behavior in the human male. Philadel-

phia: W.B. Saunders.Kramer, H. & J. Sprenger 1970 (1486). Malleus maleficarum (trans. M. Summers). (Second

edition). New York: Benjamin Blom.Kulick, D. & M.Willson 1995. Taboo: sex, identity and erotic subjectivity in anthropological fieldwork.

London: Routledge.LaBerge, S. 1985. Lucid dreaming: the power of being awake and aware in your dreams. New York:

Ballantine.——— & H. Rheingold 1990. Exploring the world of lucid dreaming. New York: Ballantine.Lancre, P. de 1982 (1613). Tableau de l’inconstance des mauvais anges et démons (ed. N.

Jacques-Chaquin). Paris: Aubier.Laplanche, J. & J.-N. Pontalis 1968. Fantasy and the origin of sexuality. International Journal of

Psychoanalysis 49, 1-18.——— & ——— 1973. The language of psychoanalysis. London: Hogarth Press.Lea, H.C. 1939. Materials toward a study of witchcraft. 3 vols. Philadelphia: University of

Pennsylvania Press.Le Goff, J. 1988. Christianity and dreams (second to seventh century). In The medieval imagi-

nation, J. Le Goff, 193-231. Chicago: University of Chicago Press.Lévi-Strauss, C. 1963. The structural study of myth. In Structural anthropology, C. Lévi-Strauss,

206-31. New York: Basic Books.Levy, R. 1984. Emotion, knowing and culture. In Culture theory: essays on mind, self, and emotion

(eds) R. Shweder & R. LeVine, 214-37. Cambridge: University Press.LiDonnici, L. 1995. The Epidaurian miracle inscriptions: text, translation and commentary. Atlanta:

Scholars Press.Long, A.A. & D.N. Sedley 1987. The Hellenistic philosophers. 2 vols. Cambridge: University

Press.Low, S. 1985. Culturally interpreted symptoms or culture-bound syndromes: a cross-cultural

review of nerves. Social Science and Medicine 21, 187-96.Mack, J. 1970. Nightmares and human conflict. Boston: Little, Brown & Co.MacNish, R. 1830. The philosophy of sleep. Glasgow: W.R. McPhun.Malinowski, B. 1962 (1923). Pioneers in the study of sex and marriage: Sigmund Freud. In

Sex, culture and myth, B. Malinowski, 114-16. New York: Harcourt, Brace & World.——— 1965 (1927). Sex and repression in savage society. Cleveland: Meridian.——— 1967. A diary in the strict sense of the term. London: Routledge & Kegan Paul.Marcus, S. 1966. The other Victorians: a study of sexuality and pornography in mid-nineteenth-century

England, New York: Basic Books.Markowitz, F. & M. Ashkenazi (eds) 1999. Sex, sexuality and the anthropologist, Urbana, Ill.:

University of Illinois Press.Matheson, T. 1998. Alien abductions, Amherst, N.Y.: Prometheus Books.Oberhelman, S. 1983. Galen, On diagnosis from dreams. Journal of the History of Medicine and Allied

Sciences 38, 36-47.Pagels, E. 1988. Adam, Eve and the serpent. New York:Vintage.Pigeaud, J. 1983.Voir, imaginer, rêver, être fous. Littérature, Médecine, Société 5, 23-53.Plato, Republic 1994. (trans. R. Waterfield). Oxford: University Press.Powell, N. 1972. Fuseli: the nightmare. New York:Viking.

CHARLES STEWART 307

Prudentius, The daily round (liber cathimerinon) 1949. (trans. H.J. Thomson). (Loeb ClassicalLibrary). London: Heinemann.

Rand, N. 1994. Introduction: renewals of psychoanalysis. In The shell and the kernel: renewals ofpsychoanalysis (eds) N. Abraham & M. Torok, 1-22. Chicago: University Press.

Ravenhill, P. 1996. Dreams and reverie: images of otherworld mates among the Baule, West Africa.Washington: Smithsonian Institution Press.

Rechtschaffen, A. 1978. The single-mindedness and isolation of dreams. Sleep 1, 97-109.Roper, L. 1994. Oedipus and the Devil. London: Routledge.Roscher, W. 1900. Ephialtes: eine pathologisch-mythologische Abhandlung über die Alpträume und

Alpdämonen des klassischen Altertums. Leipzig: Teubner. [Pan and the nightmare, trans., abridgedJ. Hillman, 1979. Irving, Tex.: Spring Publications].

Russell, J.B. 1972. Witchcraft in the Middle Ages. Ithaca: Cornell University Press.Rycroft, C. 1981. The innocence of dreams. Oxford: University Press.Schmitt, J.-C. 1999. The liminality and centrality of dreams in the medieval West. In Dream

cultures: explorations in the comparative history of dreaming (eds) D. Shulman & G. Stroumsa,274-87. New York: Oxford University Press.

Seneca, On anger 1928. In Moral essays, vol. 1 (trans. J.W. Basore). (Loeb Classical Library).London: Heinemann.

Shelley, M. 1993 (1818). Frankenstein (ed. M. Butler). Oxford: University Press.Simons, R. & C. Hughes 1985. The culture-bound syndromes: folk illnesses of psychiatric and anthro-

pological interest. Dordrecht: Reidel.Sorabji, R. 1997. Is Stoic philosophy helpful as psychotherapy?. In Aristotle and after (ed.) R.

Sorabji, 197-209. (BICS Supplement 68). London: Institute of Classical Studies.Sperber, D. 1985. Anthropology and psychology: towards an epidemiology of representations.

Man 20, 73-89.Stocking, G. 1986. Anthropology and the science of the irrational: Malinowski’s encounter with

Freudian psychoanalysis. In Malinowski, Rivers, Benedict and others: essays on culture and person-ality (ed.) G. Stocking, 13-39. Madison: University of Wisconsin Press.

Tedlock, B. 1987. Dreaming and dream research. In Dreaming: anthropological and psychologicalapproaches (ed.) B. Tedlock, 1-30. Cambridge: University Press.

——— 1994.The evidence from dreams. In Handbook of psychological anthropology (ed.) P. Bock,279-95. Westport, Conn.: Greenwood Press.

Tertullian, On the soul 1947. In De anima (ed. J.H. Waszink). Amsterdam: J.M.Meulenhoff.

Tillhagen, C.-H. 1960. The conception of the nightmare in Sweden. In Humaniora: essays in literature, folklore, bibliography (eds) W. Hand & G. Arlt, 317-29. Locust Valley, NY: J.J. Augustin.

Todorov, T. 1973. The fantastic: a structural approach to a literary genre. Cleveland: Case WesternReserve Press.

Tomory, P.A. 1972. The art and life of Henry Fuseli. New York: Praeger.Torgovnick, M. 1990. Gone primitive: savage intellects, modern lives. Chicago: University Press.von Staden, H. 1989. Herophilus: the art of medicine in early Alexandria. Cambridge: University

Press.Waller, J. 1816. Treatise on the incubus, or night-mare, disturbed sleep, terrific dreams, and nocturnal

visions: with the means of removing these distressing complaints. London: E. Cox & Son.Wayne (Malinowska), H. 1985. Bronislaw Malinowski: the influence of various women on his

life and works. American Ethnologist 12, 529-40.Weber, M. 1991 (1904-5). The Prostestant ethic and the spirit of capitalism (trans. T. Parsons).

London: HarperCollins.Webster, R. 1995. Why Freud was wrong: sex, sin and psychoanalysis. London: HarperCollins.Wengle, J. 1988. Ethnographers in the field: the psychology of research, Tuscaloosa, Al.: University of

Alabama Press.Weyer, J. 1991 (1583). Witches, devils and doctors in the Renaissance. Johann Weyer: De praestigiis dae-

monum (ed. G. Mora). Binghamton, N.Y.: Center for Medieval & Renaissance Studies.Winkler, J. 1990. Unnatural acts: erotic protocols in Artemidoros’ Dream analysis. In The con-

straints of desire, J. Winkler, 17-44. London: Routledge.Woodburn, J.C. 1964. The social organization of the Hadza of northern Tanganyika. Ph.D.

thesis, University of Cambridge.

308 CHARLES STEWART

Rêves érotiques et cauchemars depuis l’antiquité jusqu’ici

Résumé

L’histoire des rêves érotiques, des cauchemars et des cauchemars érotiques offre une occa-sion précieuse d’étudier comment de tels rêves ont mis à l’épreuve les idées occidentales surle soi, le désir et la maîtrise de soi. Comme Foucault, je considère qu’il est plus productifd’analyser ces rêves, et les difficultés de leur introjection, en tant que sites de constuction desoi plutôt qu’en termes de répression. Les rêves et cauchemars érotiques ont été infléchis parplusieurs stratégies historiques de fabrication du soi, produites elles-mêmes par différentsrégimes de savoir tels que l’ascétisme chrétien, la médecine ou la philosophie. Les cauchemarsérotiques continuent à proliférer dans les récits d’enlèvements par des extra-terrestres. Uneraison pour cette ténacité historique a été l’aise avec laquelle les sensations affectives descauchemars érotiques – terreur et excitation sexuelle – sont passées d’un genre à l’autre parmides genres aussi divers que les manuels monacaux, les contes médiévaux, la fiction gothiqueet les rêves personnels. Cette étude démontre l’importance d’une perspective historique afin de pouvoir identifier et comprendre les syndromes qui sont élaborés culturellement (ou‘syndromes culturels spécifiques’).

Dept of Anthropology, University College London WC1E 6BT. c.stewart@ucl.ac.uk

CHARLES STEWART 309